Structural determinants in addition to the amino-terminal sorting sequence influence membrane localization of Escherichia coli lipoproteins.

The lipid-modified nine-residue amino-terminal sequence of the mature form of the major outer membrane lipoprotein of Escherichia coli contains information that is responsible for sorting to either the inner or outer membrane. Fusion of this sorting sequence to beta-lactamase is sufficient for localization of the resultant lipo-beta-lactamase to the outer membrane (J. Ghrayeb and M. Inouye, J. Biol. Chem. 259:463-467, 1984). Substitution of the serine adjacent to the amino-terminal lipid-modified cysteine residue of the sorting sequence with the negatively charged residue aspartate causes inner membrane localization (K. Yamaguchi, F. Yu, and M. Inouye, Cell 53:423-432, 1988). Fusion of the aspartate-containing nine-residue inner membrane localization signal to the normally outer membrane lipoprotein bacteriocin release protein does cause partial localization to the inner membrane. However, a single replacement of the glutamine adjacent to the amino-terminal lipid-modified cysteine residue of bacteriocin release protein with aspartate causes no inner membrane localization. Therefore, an aspartate residue itself lacks the information necessary for inner membrane sorting when removed from the structural context provided by the additional eight residues of the sorting sequence. Although the aspartate-containing inner membrane sorting sequence causes an almost quantitative localization to the inner membrane when fused to the otherwise soluble protein beta-lactamase, this sequence cannot prevent significant outer membrane localization when fused to proteins (bacteriocin release protein and OmpA) normally found in the outer membrane. Therefore, structural determinants in addition to the amino-terminal sorting sequence influence the membrane localization of lipoproteins.     

Localization of the outer membrane subunit OprM of resistance-nodulation-cell division family multicomponent efflux pump in Pseudomonas aeruginosa

The outer membrane subunit OprM of the multicomponent efflux pump of Pseudomonas aeruginosa has been assumed to form a transmembrane xenobiotic exit channel across the outer membrane. We challenged this hypothesis to clarify the underlying ambiguity by manipulating the amino-terminal signal sequence of the OprM protein of the MexAB-OprM efflux pump in P. aeruginosa. [(3)H]Palmitate uptake experiments revealed that OprM is a lipoprotein. The following lines of evidence unequivocally established that the OprM protein functioned at the periplasmic space. (i) The OprM protein, in which a signal sequence including Cys-18 was replaced with that of periplasmic azurin, appeared in the periplasmic space but not in the outer membrane fraction, and the protein fully functioned as the pump subunit. (ii) The hybrid OprM containing the N-terminal transmembrane segment of the inner membrane protein, MexF, appeared exclusively in the inner membrane fraction. The hybrid protein containing 186 or 331 amino acid residues of MexF was fully active for the antibiotic extrusion, but a 42-residue protein was totally inactive. (iii) The mutant OprM, in which the N-terminal cysteine residue was replaced with another amino acid, appeared unmodified with fatty acid and was fractionated in both the periplasmic space and the inner membrane fraction but not in the outer membrane fraction. The Cys-18-modified OprM functioned for the antibiotic extrusion indistinguishably from that in the wild-type strain. We concluded, based on these results, that the OprM protein was anchored in the outer membrane via fatty acid(s) attached to the N-terminal cysteine residue and that the entire polypeptide moiety was exposed to the periplasmic space.     

Amino acids at positions 3 and 4 determine the membrane specificity of Pseudomonas aeruginosa lipoproteins

Escherichia coli lipoproteins with Asp at position 2 remain in the inner membrane, whereas those having other amino acids are targeted to the outer membrane by the Lol system. However, inner membrane lipoproteins without Asp at position 2 are found in other Gram-negative bacteria. MexA of Pseudomonas aeruginosa, an inner membrane-specific lipoprotein involved in multidrug efflux, has Gly at position 2. To identify the residue or region of MexA that functions as an inner membrane retention signal, we constructed chimeric lipoproteins comprising various regions of MexA and an outer membrane lipoprotein, OprM, and analyzed their membrane localization. Lys and Ser at positions 3 and 4, respectively, were found to be critical for the inner membrane localization of MexA in P. aeruginosa. Substitution of these residues with Leu and Ile, which are present in OprM, was sufficient to target the chimeric lipoprotein to the outer membrane and to abolish the ability of MexA to confer drug resistance. The membrane specificity of a model lipoprotein, lipoMalE, a lipidated variant of the periplasmic maltose-binding protein of E. coli, was also determined by the residues at positions 3 and 4 in P. aeruginosa. In contrast to the widely accepted "+2 rule" for E. coli lipoproteins, these results suggest a new "+3, +4 rule" for lipoprotein sorting in P. aeruginosa, namely, the final destination of lipoproteins is determined by the residues at positions 3 and 4.     

Expert system for predicting protein localization sites in gram-negative bacteria

We have developed an expert system that makes use of various kinds of knowledge organized as "if-then" rules for predicting protein localization sites in Gram-negative bacteria, given the amino acid sequence information alone. We considered four localization sites: the cytoplasm, the inner (cytoplasmic) membrane, the periplasm, and the outer membrane. Most rules were derived from experimental observations. For example, the rule to recognize an inner membrane protein is the presence of either a hydrophobic stretch in the predicted mature protein or an uncleavable N-terminal signal sequence. Lipoproteins are first recognized by a consensus pattern and then assumed present at either the inner or outer membrane. These two possibilities are further discriminated by examining an acidic residue in the mature N-terminal portion. Furthermore, we found an empirical rule that periplasmic and outer membrane proteins were successfully discriminated by their different amino acid composition. Overall, our system could predict 83% of the localization sites of proteins in our database.     

Amino acid sequence of a lysozyme (B-enzyme) from Bacillus subtilis YT-25

The amino acid sequence of a lysozyme, (B-enzyme), from Bacillus subtilis YT-25 was determined by conventional methods. B-Enzyme comprised 117 amino acid residues and had a heterogeneous sequence in the amino-terminal region. The amino acid sequence of B-enzyme was different from those of all other lysozymes the sequences of which are known. However, the partial amino acid sequence of Ser(74) to Ser(97) of B-enzyme was homologous with that of the active-site region of hen egg-white lysozyme (Ser(36) to Ser(60], which includes one of the catalytic amino acids, Asp(52). It is interesting that B-enzyme has an amino acid sequence homologous with that of the gag protein p25 of the AIDS virus ARV-2.     

The protein sequence responsible for lipoprotein membrane localization in Escherichia coli exhibits remarkable specificity

Structural information defining an N-terminal sequence required for the membrane sorting of bacterial lipoproteins has been previously garnered through the study of a hybrid outer membrane (OM) lipo-beta-lactamase (LL) (Ghrayeb and Inouye (1984) J. Biol. Chem. 259, 463-467). Introduction of an aspartate as the second residue of mature LL (D2 mutant) causes an inner membrane (IM) localization of this protein (Yamaguchi, K., Yu, F., and Inouye, M. (1988) Cell 53, 423-432). Introduction of as aspartate at the third residue of mature LL (D3) causes a weaker IM sorting signal and when present as the fourth residue (D4), normal OM sorting occurs. A positively charged residue at the second position (K2) has no effect on OM localization. Remarkably, glutamate substitution at either the second (E2) or third (E3) position does not interfere with OM sorting. Sorting of the mutant D2 LL can be partially suppressed by introduction of a positively charged histidine (D2H3) or lysine (D2K3) at residue 3 of the mature protein. These results indicate that both the negative charge of the aspartate residue and some structural feature not present in a glutamate residue are required for sorting to the IM. The suppression of IM localization of the D2H3 LL double mutant can be eliminated by growing Escherichia coli at pH 8.4 to reduce the histidine partial positive charge. This result supports the essentiality of a negative charge in IM localization and indicates that the committed step in lipoprotein sorting is made in a cellular compartment, the periplasm, at equilibrium with the external pH.     

The complete amino acid sequence of rat submaxillary gland tonin does contain the aspartic acid at the active site: confirmation by protein sequence analysis

The revised amino acid sequence of rat submaxillary gland tonin, a serine protease, does contain the active site Asp residue. The active site of this kallikrein-related enzyme is thus made up of the same catalytic triad (Asp, Ser, and His) found in all known serine proteases. The important Asp residue has now been localized in a 16 amino acid peptide previously reported as missing in the tonin sequence. The complete amino acid sequence thus contains 235 residues corresponding to a molecular weight of 25,658, more in agreement with previously reported molecular weights. Moreover, the revised structure led (a) to the assignment of Arg, Asn, and Val residues instead of His, Asp, and Gly at positions 63, 165, and 169, respectively; (b) to the assignment of residues occupying an overlapping sequence at positions 165-171, and finally (c) to the localization of two N-glycosylation sites at positions 82 and 165. These results further document the close relationship of tonin to the ever expanding kallikrein family.     

A CLN2-related and thermostable serine-carboxyl proteinase, kumamolysin: cloning, expression, and identification of catalytic serine residue

The gene encoding kumamolysin, a thermostable pepstatin-insensitive carboxyl proteinase, was cloned and expressed. (i) Kumamolysin was synthesized as a large precursor consisting of two regions: amino-terminal prepro (188 amino acids) and mature proteins (384 amino acids). (ii) The deduced amino acid sequence of the mature region exhibited high similarity to those of such bacterial pepstatin-insensitive enzymes as Pseudomonas carboxyl proteinase (PSCP; EC 3.4.23.37, identity = 37%), Xanthomonas carboxyl proteinase (XCP; EC 3.4.23.33, identity = 36%), and human CLN2 gene product (identity = 36%), which is related to a fatal neurodegenerative disease. (iii) The presumed catalytic triad, Glu78, Asp82, Ser278 [three-dimensional structure of PSCP: Wlodawer, A. et al. (2001) Nature Struct. Biol., 8, 442-446], was found to be conserved in the amino acid sequence of kumamolysin. (iv) Kumamolysin was inactivated by such aldehyde-type inhibitors as Ac-Ile-Pro-Phe-CHO (K(i) = 0.7 0.14 microM). In PSCP, it has been clarified that these inhibitors form a hemiacetal linkage with the catalytic serine residue and inactivate the enzyme. (v) Mutational analysis of the Ser278 residue revealed that the mutant lost both auto-processing activity and proteolytic activity. These results strongly suggest that kumamolysin has a unique catalytic triad consisting of Glu78, Asp82, and Ser278 residues, as previously observed for PSCP.     

Precise determination of the mitochondrial import signal contained in a 70 kDa protein of yeast mitochondrial outer membrane

A major 70 kDa protein of the yeast mitochondrial outer membrane is coded by a nuclear gene, synthesized on cytoplasmic ribosomes, and transported to the mitochondrial outer membrane. In order to investigate in detail the information necessary for localizing the 70 kDa protein at the outer membrane, we have examined the intracellular and intramitochondrial location of fusion proteins which consist of various lengths of the amino-terminal region of the 70 kDa protein with an enzymatically active beta-galactosidase. The results indicate that the extreme amino-terminal 12 amino acids of the 70 kDa protein function as a targeting sequence, whereas the subsequent uncharged region (up to residue 29) is necessary for "stop-transfer" and "anchoring" functions. Moreover, we have found that a fusion protein which contained the amino-terminal 19 amino acids of the 70 kDa protein is localized on the outer membrane as well as in the matrix space. Changes in the dual localization of this fusion protein accompanied its overproduction or expression in a respiration-deficient yeast mutant.     

Effects of inserting eight amino acid residues into the major lipoprotein on its assembly in the outer membrane of Escherichia coli.

A DNA sequence consisting of 24 base pairs was inserted into the structural gene (lpp) coding for the major lipoprotein of the Escherichia coli outer membrane which was carried on a high-copy-number plasmid in which expression was regulated through a lac promoter-operator region. This modification resulted in the insertion of eight amino acid residues, Glu-Glu-Phe-Leu-Glu-Glu-Phe-Leu, between the glutamine residue at position 9 and the leucine residue at position 10 of the wild-type lipoprotein sequence. When production of the mutant lipoprotein was induced by a lac inducer, the cells became swollen, showed unusual morphology, and eventually lysed. When the membrane fraction was analyzed after the induction, the mutant lipoprotein was found to have been normally secreted across the cytoplasmic membrane and assembled in the outer membrane. This lipoprotein was modified with glycerol and palmitic acid and even formed the bound form, which was linked covalently to peptidoglycan. The major difference between the membrane-associated mutant lipoprotein and the wild-type lipoprotein was that the mutant lipoprotein became sensitive to trypsin treatment. These results indicate that the substantial alteration in mutant lipoprotein structure near the amino-terminal end does not interfere with modification of the amino-terminal cysteine residue or cleavage of the signal peptide by the prolipoprotein-specific signal peptidase. However, this mutant lipoprotein assembled in the outer membrane appears to have deleterious effects with respect to envelope structure and cellular morphology and viability.     

Amino acid residue penultimate to the amino-terminal gly residue strongly affects two cotranslational protein modifications, N-myristoylation and N-acetylation

To examine the amino-terminal sequence requirements for cotranslational protein N-myristoylation, a series of site-directed mutagenesis of N-terminal region were performed using tumor necrosis factor as a nonmyristoylated model protein. Subsequently, the susceptibility of these mutants to protein N-myristoylation was evaluated by metabolic labeling in an in vitro translation system or in transfected cells. It was found that the amino acid residue at position 3 in an N-myristoylation consensus motif, Met-Gly-X-X-X-Ser-X-X-X, strongly affected the susceptibility of the protein to two different cotranslational protein modifications, N-myristoylation and N-acetylation; 10 amino acids (Ala, Ser, Cys, Thr, Val, Asn, Leu, Ile, Gln, and His) with a radius of gyration smaller than 1.80 A directed N-myristoylation, two negatively charged residues (Asp and Glu) directed N-acetylation, and two amino acids (Gly and Met) directed heterogeneous modification with both N-myristoylation and N-acetylation. The amino acid requirements at this position for the two modifications were dramatically changed when Ser at position 6 in the consensus motif was replaced with Ala. Thus, the amino acid residue penultimate to the N-terminal Gly residue strongly affected two cotranslational protein modifications, N-myristoylation and N-acetylation, and the amino acid requirements at this position for these two modifications were significantly affected by downstream residues.     

Molecular cloning and sequence analysis of cDNA for batroxobin, a thrombin-like snake venom enzyme

Determination of the nucleotide sequence of a cDNA for batroxobin, a thrombin-like enzyme from Bothrops atrox, moojeni venom, allowed elucidation of the complete amino acid sequence of batroxobin for the first time for a thrombin-like snake venom enzyme. The molecular weight of batroxobin is 25,503 (231 amino acids). The amino acid sequence of batroxobin exhibits significant homology with those of mammalian serine proteases (trypsin, pancreatic kallikrein, and thrombin), indicating that batroxobin is a member of the serine protease family. Based on this homology and enzymatic and chemical studies, the catalytic residues and disulfide bridges of batroxobin were deduced to be as follows: catalytic residues, His41, Asp86, and Ser178; and disulfide bridges, Cys7-Cys139, Cys26-Cys42, Cys74-Cys230, Cys118-Cys184, Cys150-Cys163, and Cys174-Cys199. The amino-terminal amino acid residue of batroxobin, valine, is preceded by 24 amino acids. This may indicate that the amino-terminal hydrophobic peptide (18 amino acids) is a prepeptide and that the hydrophilic peptide (6 amino acids), preceded by the putative prepeptide, is a propeptide.     

Mechanism underlying the inner membrane retention of Escherichia coli lipoproteins caused by Lol avoidance signals

Escherichia coli lipoproteins are localized to either the inner or outer membrane depending on the residue at position 2. The inner membrane retention signal, Asp at position 2 in combination with certain residues at position 3, functions as a Lol avoidance signal, i.e. the signal inhibits the recognition of lipoproteins by LolCDE that releases lipoproteins from the inner membrane. To understand the role of the residue at position 2, outer membrane-specific lipoproteins with Cys at position 2 were subjected to chemical modification followed by the release reaction in reconstituted proteoliposomes. Sulfhydryl-specific introduction of nonprotein molecules or a negative charge to Cys did not inhibit the LolCDE-dependent release. In contrast, oxidation of Cys to cysteic acid resulted in generation of the Lol avoidance signal, indicating that the Lol avoidance signal requires a critical length of negative charge at the second residue. Furthermore, not only modification of the carboxylic acid of Asp at position 2 but also that of the amine of phosphatidylethanolamine abolished the Lol avoidance function. Based on these results, the Lol avoidance mechanism is discussed.     

Direct visualization of red fluorescent lipoproteins indicates conservation of the membrane sorting rules in the family Enterobacteriaceae

Chimeras created by fusing the monomeric red fluorescent protein (RFP) to a bacterial lipoprotein signal peptide (lipoRFPs) were visualized in the cell envelope by epifluorescence microscopy. Plasmolysis of the bacteria separated the inner and outer membranes, allowing the specific subcellular localization of lipoRFPs to be determined in situ. When equipped with the canonical inner membrane lipoprotein retention signal CDSR, lipoRFP was located in the inner membrane in Escherichia coli, whereas the outer membrane sorting signal CSSR caused lipoRFP to localize to the outer membrane. CFSR-RFP was also routed to the outer membrane, but CFNSR-RFP was located in the inner membrane, consistent with previous data showing that this sequence functions as an inner membrane retention signal. These four lipoproteins exhibited identical localization patterns in a panel of members of the family Enterobacteriaceae, showing that the lipoprotein sorting rules are conserved in these bacteria and validating the use of E. coli as a model system. Although most predicted inner membrane lipoproteins in these bacteria have an aspartate residue after the fatty acylated N-terminal cysteine residue, alternative signals such as CFN can and probably do function in parallel, as indicated by the existence of putative inner membrane lipoproteins with this sequence at their N termini.     

Sequences required for delivery and localization of the ADP/ATP translocator to the mitochondrial inner membrane.

The ADP/ATP translocator, a transmembrane protein of the mitochondrial inner membrane, is coded in Saccharomyces cerevisiae by the nuclear gene PET9. DNA sequence analysis of the PET9 gene showed that it encoded a protein of 309 amino acids which exhibited a high degree of homology with mitochondrial translocator proteins from other sources. This mitochondrial precursor, in contrast to many others, does not contain a transient presequence which has been shown to direct the posttranslational localization of proteins in the organelle. Gene fusions between the PET9 gene and the gene encoding beta-galactosidase (lacZ) were constructed to define the location of sequences necessary for the mitochondrial delivery of the ADP/ATP translocator protein in vivo. These studies reveal that the information to target the hybrid molecule to the mitochondria is present within the first 115 residues of the protein. In addition, these studies suggest that the "import information" of the amino-terminal region of the ADP/ATP translocator precursor is twofold. In addition to providing targeting function of the precursor to the organelle, these amino-terminal sequences act to prevent membrane-anchoring sequences located between residues 78 and 98 from stopping import at the outer mitochondrial membrane. These results are discussed in light of the function of distinct protein elements at the amino terminus of mitochondrially destined precursors in both organelle delivery and correct membrane localization.     

Protein sorting between the outer and inner mitochondrial membranes: submitochondrial localization of cytochrome c1 whose presequence is replaced by the amino-terminal region of a 70 kDa outer membrane protein

The amino-terminal region of a 70 kDa mitochondrial outer membrane protein of yeast and the presequence of cytochrome c1, an inner membrane protein exposed to the intermembrane space, are thought to be responsible for localizing the proteins in their final destinations after synthesis in the cytosol. Gene fusion experiments were used to identify signals that are responsible for protein sorting between the outer and inner mitochondrial membranes. The submitochondrial localization of cytochrome c1 whose presequence was replaced by the amino-terminal region of the 70 kDa mitochondrial outer membrane protein has been investigated. We have also used an in vivo complementation assay to determine whether or not a 70k-cyt c1 fusion protein is functional. Both the first half and all of the presequence of cytochrome c1 can be replaced by the amino-terminal 12 or 29 residues of the 70 kDa protein for transport to the inner membrane and functional assembly into succinate-cytochrome c reductase. However, replacements by the amino-terminal 61 residues of the 70 kDa protein result in exclusive localization of the fusion proteins to the outer membrane, and the fusions cannot be assembled into the enzyme complex. These data indicate that a mitochondrial targeting signal alone is sufficient to direct cytochrome c1 of mature size to the inner membrane.     

Nucleotide sequence of the gene for the ferrienterochelin receptor FepA in Escherichia coli. Homology among outer membrane receptors that interact with TonB

We have determined the nucleotide sequence of the Escherichia coli fepA gene, which codes for the outer membrane receptor for ferrienterochelin and colicins B and D. The predicted FepA polypeptide has a molecular weight of 79,908 and consists of 723 amino acids. A 22-amino acid leader or signal peptide preceded the mature protein. With respect to overall composition, hydropathy, net charge and distribution of nonpolar segments, the FepA polypeptide was typical of other E. coli outer membrane proteins, except that FepA contained 2 cysteine residues. Comparison of the deduced amino acid sequence of FepA with that of three other TonB-dependent receptors (BtuB, FhuA, and IutA) revealed only a few regions of sequence homology; one of these included the amino-termini. An amino acid substitution within the conserved amino-terminal region of BtuB resulted in production of a receptor that had normal binding functions but was incapable of energy-dependent transport of vitamin B12. This result suggests that the amino-terminal end of these four polypeptides is involved in interaction with the TonB protein or another step of energy transduction. Three other regions of homology were shared among the four proteins: one near residues 50 to 70, another at about residue 100 to 140, and the last between 20 and 40 amino acid residues from the carboxyl terminus. The function of these three regions remains speculative.     

Amino acid sequence of human plasma apolipoprotein C-II from normal and hyperlipoproteinemic subjects

The complete amino acid sequence of human plasma apolipoprotein C-II isolated from normal individuals and a subject with type V hyperlipoproteinemia has been determined. Apo-C-II contains 79 amino acids with the following amino acid composition: Asp4, Asn1, Thr9, Ser9, Glu7, Gln7, Pro4, Gly2, Ala6, Val4, Met2, Ile1, Leu8, Tyr5, Phe2, Lys6, Arg1, Trp1. Cleavage of apo-C-II by cyanogen bromide produced three peptides designated as CB-1 (9 residues), CB-2 (51 residues), and CB-3 (19 residues). Two peptides, CT-1 (50 residues) and CT-2 (29 residues), which overlapped the cyanogen bromide peptides, were obtained by tryptic digestion of citraconylated apo-C-II at the single arginine residue. The amino acid sequence of apo-C-II was obtained by the automated phenyl isothiocyanate degradation of intact apo-C-II and the peptides produced by cleavage of apo-C-II by cyanogen bromide and trypsin. Phenylthiohydantoins were identified by high performance liquid chromatography and chemical ionization-mass spectrometry. The amino acid sequence of apo-C-II from the normal subject was identical with the apo-C-II isolated from the hyperlipoproteinemic subject.     

A novel outer membrane lipoprotein, LolB (HemM), involved in the LolA (p20)-dependent localization of lipoproteins to the outer membrane of Escherichia coli.

The Escherichia coli major outer membrane lipoprotein (Lpp) is released from the inner membrane into the periplasm as a complex with a carrier protein, LolA (p20), and is then specifically incorporated into the outer membrane. An outer membrane protein playing a critical role in Lpp incorporation was identified, and partial amino acid sequences of the protein, named LolB, were identical to those of HemM, which has been suggested to play a role in 5-aminolevulinic acid synthesis in the cytosol. In contrast to this suggested role, the deduced amino acid sequence of HemM implied that the gene encodes a novel outer membrane lipoprotein. Indeed, an antibody raised against highly purified LolB revealed its outer membrane localization, and inhibited in vitro Lpp incorporation into the outer membrane. Furthermore, LolB was found to be synthesized as a precursor with a signal sequence and then processed to a lipid-modified mature form. An E.coli strain possessing chromosomal hemM under the control of the lac promoter-operator required IPTG for growth, indicating that hemM (lolB) is an essential gene. Outer membrane prepared from LolB-depleted cells did not incorporate Lpp. When the Lpp-LolA complex was incubated with a water-soluble LolB derivative, Lpp was transferred from LolA to LolB. Based on these results, the outer membrane localization pathway for E.coli lipoprotein is discussed with respect to the functions of LolA and LolB.     

Amino acid substitutions in pilin of Pseudomonas aeruginosa. Effect on leader peptide cleavage, amino-terminal methylation, and pilus assembly.

A total of 37 separate mutants containing single and multiple amino acid substitutions in the leader and amino-terminal conserved region of the Type IV pilin from Pseudomonas aeruginosa were generated by oligonucleotide-directed mutagenesis. The effect of these substitutions on the secretion, processing, and assembly of the pilin monomers into mature pili was examined. The majority of substitutions in the highly conserved amino-terminal region of the pilin monomer had no effect on piliation. Likewise, substitution of several of the residues within the six amino acid leader sequence did not affect secretion and leader cleavage (processing), including replacement of one or both of the positively charged lysine residues with uncharged or negatively charged amino acids. One characteristic of the Type IV pili is the presence of an amino-terminal phenylalanine after leader peptide cleavage which is N-methylated prior to assembly of pilin monomers into pili. Substitution of the amino-terminal phenylalanine with a number of other amino acids, including polar, hydrophobic, and charged residues, did not affect proper leader cleavage and subsequent assembly into pili. Amino-terminal sequencing showed that the majority of substitute residues were also methylated. Substitution of the glycine residue at the -1 position to the cleavage site resulted in the inability to cleave the prepilin monomers and blocked the subsequent assembly of monomers into pili. These results indicate that despite the high degree of conservation in the amino-terminal sequences of the Type IV pili, N-methylphenylalanine at the +1 position relative to the leader peptide cleavage site is not strictly required for pilin assembly. N-Methylation of the amino acids substituted for phenylalanine was shown to have taken place in four of the five mutants tested, but it remains unclear as to whether pilin assembly is dependent on this modification. Recognition and proper cleavage of the prepilin by the leader peptidase appears to be dependent only on the glycine residue at the -1 position. Cell fractionation experiments demonstrated that pilin isolated from mutants deficient in prepilin processing and/or assembly was found in both inner and outer membrane fractions, indistinguishable from the results seen with the wild type.